CN209230716U - A kind of volume measurement device - Google Patents
A kind of volume measurement device Download PDFInfo
- Publication number
- CN209230716U CN209230716U CN201920224655.5U CN201920224655U CN209230716U CN 209230716 U CN209230716 U CN 209230716U CN 201920224655 U CN201920224655 U CN 201920224655U CN 209230716 U CN209230716 U CN 209230716U
- Authority
- CN
- China
- Prior art keywords
- camera
- central processing
- processing unit
- depth
- test desk
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The utility model proposes a kind of volume measurement devices, including central processing unit and test desk；The test desk includes vision camera unit, dynamic weighing sensor；The vision camera unit is mainly made of the depth camera for being directed at the multiple angles in imaging area；The depth camera is connect with central processing unit；Dynamic weighing sensor is placed in below imaging area, and acquisition weight data is sent to central processing unit.The utility model is that the functional requirement of " by more depth camera calibration techniques and point cloud matching technology; can merge the depth data at multiple visual angles; the three-dimensional space information for reconstructing irregularly shaped object improves the stability, robustness and measurement accuracy of measuring unit " provides hardware configuration；Hardware configuration is provided for the technical solution of " three-dimensional space of irregularly shaped object being cut into the irregular figure of n piece using microtomy, the volume of irregularly shaped object is sought by integration method ".
Description
Technical field
The utility model belongs to automation equipment field, especially relates to a kind of volume measurement device.
Background technique
Gangue is the solid waste discharged in coal process of manufacture, is the rock that a kind of phosphorus content is lower, harder than coal
Stone.In order to improve the quality of coal, gangue sorting is a link indispensable in coal production.
Gangue sorting at present mainly has wet cleaning method, dry separation method and manual sorting's method.Wet cleaning method includes jigging method, dense medium
Method, its main feature is that it is different from gangue density using coal, raw coal is placed in solution, realizes the separation of coal and gangue.It is such
Equipment needed for method is huge, and serious environmental pollution is caused in complex process and sorting efficiency underground.Dry separation method including the use of coal with
Both gangues intensity is different and uses the broken choosing of roller, or is known to light transmission difference using dual energy gamma ray projection using the two
It does not sort, the former destroys coal cinder shape, and sorting rate is poor, and equipment is huge, and the latter's facility is expensive, and there is radiation.In addition, people
Choosing is easily falsely dropped and leaked to work screening operation bad environments, large labor intensity.
In order to realize the automatically screening of coal and gangue, can be calculated by the cubing to material, density, thus
It identifies coal and gangue, therefore, intelligentized cubing is carried out for material, especially seeks the volume of irregularly shaped object,
The problem of as current urgent need to resolve.
Utility model content
The utility model provides a kind of volume measurement device, can be " to seek the volume of irregularly shaped object, be greatly improved
The technical solution of volume computational accuracy, and adaptability to degree of irregularity " provides hardware configuration.
In order to achieve the above objectives, the technical solution of the utility model is achieved in that
A kind of volume measurement device, including central processing unit and test desk；The test desk includes vision camera unit, moves
State weighing sensor；The vision camera unit is mainly made of the depth camera for being directed at the multiple angles in imaging area；The depth
Degree camera is connect with central processing unit；Dynamic weighing sensor is placed in below imaging area, during acquisition weight data is sent to
Central processor.
Further, the depth camera is uniformly distributed around the determinand in 360 degree of regions.
Further, it is additionally provided with camera shutter trigger, the camera shutter trigger is set to test desk inlet side
The position of edge.
Further, it is additionally provided with sorting executing agency, the sorting executing agency includes scraping wings and hopper, wherein pusher
Plate is arranged on test desk, connect with central processing unit；The test desk side in scraping wings pusher direction is arranged in the hopper.
Compared with prior art, the utility model have it is following the utility model has the advantages that
(1) the utility model is that " by more depth camera calibration techniques and point cloud matching technology, can merge multiple visual angles
Depth data, reconstruct the three-dimensional space information of irregularly shaped object, improve the stability, robustness and measurement of measuring unit
Precision；" functional requirement necessary hardware configuration is provided；
(2) the utility model is " three-dimensional space of irregularly shaped object to be cut into the irregular figure of n piece using microtomy
Shape seeks the volume of irregularly shaped object by integration method, greatly improves volume computational accuracy, and fit to degree of irregularity
Ying Xing；" technical solution necessary hardware configuration is provided；
(3) the utility model uses depth camera photographic subjects object, obtains object volume, eliminates the shadow of natural lighting
It rings, is not required to add secondary light source in image acquisition process；
(4) object being measured does not need 360 degree rotation in the utility model, improves the adaptation of more measurement environment and equipment
Property, simpler is become to image acquisition units apparatus structure；
(5) it is measured in the utility model and only needs 3 frame depth camera datas, improved the real-time of measurement, meet flowing water
The requirement of line on-line measurement.
Detailed description of the invention
Fig. 1 is the structural schematic diagram one of the test desk of the utility model embodiment；
Fig. 2 is the structural schematic diagram two of the test desk of the utility model embodiment.
Wherein:
13, correlation photoelectric sensor；14, depth camera；
15, scraping wings；16, hopper.
Specific embodiment
It should be noted that in the absence of conflict, the feature in the embodiments of the present invention and embodiment can
To be combined with each other.
As shown in Figure 1 and Figure 2, in the utility model, camera shutter trigger is a pair of of correlation photoelectric sensor 13, is set
In test desk close on the rolling wiring at shooting area edge；The correlation photoelectric sensor 13 connects central processing unit；
Test desk is made of vision camera unit, the dynamic weighing sensor of shooting area；
Vision camera unit is made of the depth camera 14 for being located at the multiple angles of object under test, and depth camera 14 can choose
TOF camera, Kinect camera；Depth camera 14 is connect with central processing unit, is controlled by central processing unit；
When sample enters in test desk shooting area, correlation photoelectric sensor 13 sends trigger signal to central processing
Device, central processing unit controlling depth camera 14 acquire image, interfere with each other to avoid camera from taking pictures, cannot shoot, need simultaneously
Separated in time is taken pictures respectively by depth camera 14；
Dynamic weighing sensor is placed in lower section, the weight data of acquisition in imaging area and is sent to central processing unit；
Executing agency's setting is sorted in test desk, including scraping wings 15 and hopper 16, the wherein setting of scraping wings 15 is measuring
On platform 5, it is connect with central processing unit；5 side of test desk in 15 pusher direction of scraping wings is arranged in the hopper 16.
Cubing step are as follows:
(1) calibration of single depth camera 14
Commercial depth camera 14 (by taking TOF camera Microsoft Kinect as an example), generally all comprising a RGB camera shooting
The infrared camera of head and a perceived depth, therefore the calibration between single depth camera 14 is related to RGB camera and red
The calibration of outer camera internal reference and and two cameras between outer ginseng calibration.
Internal reference calibration uses famous Zhang Zhengyou calibration method, the black and white chessboard of multiple different positions and poses of fixed camera station acquisition
The image (paying attention to the active infrared light source for needing to shelter from depth camera 14 when infrared camera acquisition) of case marker fixed board, using such as
Lower formula calculates the Intrinsic Matrix of camera:
Wherein, M is the Intrinsic Matrix of camera, and dimension is 3 × 3；R is the outer parameter spin matrix of camera, dimension is 3 ×
3；T is the outer parameter translational vector of camera, and dimension is 3 × 1, [X Y Z]TFor the point coordinate under world coordinate system；[u v]TFor figure
As the point coordinate under coordinate system；S is zoom factor, generally equivalent to the depth information under camera coordinates system.
Outer ginseng calibration is actually the relative pose calibration between RGB camera and infrared camera, RGB camera and red
Position orientation relation between outer camera can indicate are as follows:
Wherein PirFor the point coordinate under infrared camera coordinate system, dimension is 3 × 1；PrgbFor under RGB camera coordinate system
Point coordinate, dimension be 3 × 1；For infrared camera coordinate system to the spin matrix of RGB camera coordinate system, dimension 3
×3；For infrared camera coordinate system to the translation matrix of RGB camera coordinate system, dimension is 3 × 1.
Similarly, the transformational relation of point to the RGB camera coordinate system and infrared camera coordinate system under world coordinate system can
To indicate are as follows:
WhereinFor world coordinate system to the spin matrix of RGB camera coordinate system, dimension is 3 × 3；PwFor world's seat
Point coordinate under mark system, dimension are 3 × 1；For world coordinate system to the translation matrix of RGB camera coordinate system, dimension 3
×1；For world coordinate system to the spin matrix of infrared camera coordinate system, dimension is 3 × 3；It is arrived for world coordinate system
The translation matrix of infrared camera coordinate system, dimension are 3 × 3.
It is assumed that it is identical point that RGB camera and infrared camera, which are seen, then bringing formula (3) (4) into formula (2) can
To obtain:
It is obtained after expansion:
It can further be obtained by equation (6):
RGB camera and infrared camera observe the same scaling board simultaneously, can be evaluated whether using Zhang Zhengyou calibration algorithm
The outer parameter matrix for the world coordinate system that RGB camera and infrared camera are defined relative to current scaling board out, i.e., By the observed result of multiframe bring into formula (8) can maximum likelihood estimation go outWithThat is RGB camera shooting
Rotational translation matrix between head and infrared camera.
By formula (1) it is found that point coordinate P under RGB camera coordinate systemrgbBe mapped to image coordinate system can be by following
Formula obtains:
Wherein, srgbThe depth information provided for RGB camera.
It is similarly available:
Wherein sirThe depth information provided for infrared camera.
Formula (11) is brought into formula (2), available:
(12), which are substituted into (9), to be obtained,
This formula is the formula that depth information is registered to RGB image, i.e. re-projection formula.It is carried out using multiframe observed result
Re-projection optimizes the stated accuracy that can further improve depth camera 14.
(2) calibration of multiple depth cameras 14
The target that multiple depth cameras 14 are demarcated is the coordinate system obtained between each depth camera 14 by scaling method
Transformation matrix.Under the premise of completing the calibration of single depth camera 14, what more depth cameras 14 were demarcated is in the nature the multiple depths of calibration
The transformation matrix between the RGB camera coordinate system of camera 14 is spent, scaling method is similar with above-mentioned derivation, can also use
Re-projection optimization improves the stated accuracy between multiple cameras；
The transformation matrix method particularly includes:
The outer parameter matrix characterization of camera is pose of the camera relative to some scaling board；The calibration of multiple cameras is then
According to this principle, by obtaining outer parameter matrix of multiple cameras relative to same scaling board, to obtain each depth phase
Coordinate system transformation matrix between machine 14；
(3) object depth information obtains
This method needs to obtain in 360 ° of panoramic ranges of object being measured to realize the cubing to irregularly shaped object
Depth information.It is interfered with each other between multiple depth cameras 14 to meet above-mentioned requirements and avoiding, realizes that conveyer belt object is dynamic
State measurement, the depth camera 14 of different angle needs the starting shooting of setting time interval around object under test.
When object passes through shooting area on conveyer belt, shutter trigger triggers corresponding depth camera after detecting object
14 adopt figure.Then according to the calibration result between multiple depth cameras 14, the depth information that each camera acquires can be transformed into
Under the same coordinate system, and then the panorama depth information of object is obtained.
In the present embodiment, what object depth information obtained method particularly includes: 3 depths of different angle (0 °, 120 °, 240 °)
Degree camera needs the starting shooting of setting time interval to adopt figure, and then according to the calibration result between each depth camera, (coordinate system turns
Change matrix), 3 depth maps that 3 cameras acquire are transformed under the same coordinate system, and then obtain the panorama depth letter of object
Breath；
(4) depth information pre-processes
Object data is not only contained by the depth data that multiple depth cameras 14 obtain, but also contains the backgrounds such as conveyer belt
Data, while further comprising noise data caused by environmental disturbances or equipment itself precision reason.Therefore pretreated target
To reject background data and noise data by effective method, effective testee depth information data are obtained.Relative to
Object data, background data are essentially the data of certain depth, such as conveyer belt plane.Therefore it can be incited somebody to action by dimension filter
Most of background data proposes.Later according between cloud distribution and relative geometrical relation carry out Gaussian smoothing filter then can will
Most of noise data proposes.That is the octree of building description point cloud spatial relation, so as to the quick-searching to cloud；Root
According to the distribution and relative geometrical relation between K point of arest neighbors, the mean value and variance of gauss hybrid models are calculated, it is flat to carry out Gauss
Sliding filtering proposes noise data；
(5) testee cubing
This method is directed to the cubing of irregularly shaped object, and the method that simple length, width and height cannot be used to be multiplied calculates
Volume.In general obtain point cloud be all it is sparse, the continuous profile edge of object must be obtained in order to calculate volume, needed
It is obtained using the method for fitting.Because the scrambling of body surface will necessarily be generated with plane or the method for surface fitting
Biggish error and computational efficiency it is relatively low under.This method uses for reference the thought of multiple integral, and irregularly shaped object is tieed up according to one
Degree is sliced, and irregular size is calculated to the calculating for being converted into irregular geometric figures area.It is dilute in view of point cloud data
Property is dredged, this method obtains the continuous profile of object particular slice first with local polynomial fitting method: suppose there is N number of discrete
Profile point Pi(xi,yi), i=1,2 ..., N, the polynomial function fitted by these profile points are as follows:
Y (x)=a0+a1x+a2x2+a3x3；
Wherein a is multinomial coefficient；
The objective function of construction fitting is sum of the distance of N number of discrete profile point to above-mentioned polynomial curve, and formula is as follows:
Wherein yiFor the ordinate of discrete profile point；
It is inverse by descriptor matrix, it can directly find out the optimal solution of the polynomial function y (x) of fitting；
Then the area of object particular slice, that is, the polynomial function y (x) being fitted and x-axis are calculated using the method for integral
In [xmin, xmax] area that surrounds of segment, integral formula of quadraturing is as follows:
Finally using the method for integral, the volume of determinand is calculated.The volume of determinand is object slice area in z-axis
Integral, calculation formula is as follows:
The above is only the preferred embodiment of the utility model, it is noted that for the common skill of the art
Art personnel can also make several improvements and modifications without departing from the concept of the premise utility, these improvements and modifications
Also it should be regarded as in scope of protection of the utility model.
Claims (4)
1. a kind of volume measurement device, which is characterized in that the measuring device includes central processing unit and test desk；The measurement
Platform includes vision camera unit, dynamic weighing sensor；The vision camera unit is mainly by being directed at the multiple angles in imaging area
Depth camera composition；The depth camera is connect with central processing unit；Dynamic weighing sensor is placed in below imaging area,
Acquisition weight data is sent to central processing unit.
2. volume measurement device according to claim 1, which is characterized in that the depth camera is 360 around determinand
Degree is uniformly distributed in region.
3. volume measurement device according to claim 1, which is characterized in that it is additionally provided with camera shutter trigger, it is described to take the photograph
As shutter trigger is set to the position at test desk inlet edge.
4. volume measurement device according to claim 1, which is characterized in that be additionally provided with sorting executing agency, the sorting
Executing agency includes scraping wings and hopper, and wherein scraping wings is arranged on test desk, is connect with central processing unit；The hopper is set
Set the test desk side in scraping wings pusher direction.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920224655.5U CN209230716U (en) | 2019-02-22 | 2019-02-22 | A kind of volume measurement device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201920224655.5U CN209230716U (en) | 2019-02-22 | 2019-02-22 | A kind of volume measurement device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN209230716U true CN209230716U (en) | 2019-08-09 |
Family
ID=67511181
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201920224655.5U Active CN209230716U (en) | 2019-02-22 | 2019-02-22 | A kind of volume measurement device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN209230716U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111060005A (en) * | 2019-11-28 | 2020-04-24 | 中海油能源发展股份有限公司 | Rock core volume automatic measuring device and method based on vision |
| WO2023068762A1 (en) * | 2021-10-18 | 2023-04-27 | (주)딥인사이트 | Method and device for acquiring information about solid particulate material having fluidity loaded in space having predetermined shape |
-
2019
- 2019-02-22 CN CN201920224655.5U patent/CN209230716U/en active Active
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111060005A (en) * | 2019-11-28 | 2020-04-24 | 中海油能源发展股份有限公司 | Rock core volume automatic measuring device and method based on vision |
| WO2023068762A1 (en) * | 2021-10-18 | 2023-04-27 | (주)딥인사이트 | Method and device for acquiring information about solid particulate material having fluidity loaded in space having predetermined shape |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109632033A (en) | A kind of apparatus and method for of cubing | |
| CN109848073A (en) | A kind of apparatus and method for sorting coal and gangue | |
| CN102135236B (en) | Automatic non-destructive testing method for internal wall of binocular vision pipeline | |
| WO2018028103A1 (en) | Unmanned aerial vehicle power line inspection method based on characteristics of human vision | |
| CN104713885B (en) | A kind of structure light for pcb board on-line checking aids in binocular measuring method | |
| CN105279372B (en) | A kind of method and apparatus of determining depth of building | |
| CN106969706A (en) | Workpiece sensing and three-dimension measuring system and detection method based on binocular stereo vision | |
| CN110142785A (en) | A kind of crusing robot visual servo method based on target detection | |
| CN104482860B (en) | Fish morphological parameters self-operated measuring unit and method | |
| CN108335350A (en) | The three-dimensional rebuilding method of binocular stereo vision | |
| CN106340009B (en) | A kind of electric power line detecting method and system based on parallel binocular | |
| CN106856002A (en) | A kind of unmanned plane shooting image quality evaluating method | |
| CN110648364B (en) | Multi-dimensional space solid waste visual detection positioning and identification method and system | |
| CN209230716U (en) | A kind of volume measurement device | |
| CN106679567A (en) | Contact net and strut geometric parameter detecting measuring system based on binocular stereoscopic vision | |
| CN106996748A (en) | A kind of wheel footpath measuring method based on binocular vision | |
| CN103795935B (en) | A kind of camera shooting type multi-target orientation method and device based on image rectification | |
| CN108209926A (en) | Human Height measuring system based on depth image | |
| CN106908453A (en) | The detection method and detection means of a kind of printed substrate | |
| CN109461206A (en) | A kind of the face three-dimensional reconstruction apparatus and method of multi-view stereo vision | |
| CN106295695B (en) | A kind of takeoff and landing process automatic tracing image pickup method and device | |
| CN106651957A (en) | Monocular vision target space positioning method based on template | |
| CN110047111B (en) | Parking apron corridor bridge butt joint error measuring method based on stereoscopic vision | |
| CN209520073U (en) | A kind of equipment sorting coal and gangue | |
| CN105741275B (en) | A kind of people's vehicle target's feature-extraction method based on fixed camera automatic Calibration |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |